AI Video Prompts for Fashion
Runway walks, editorial close-ups, fabric in motion, and the light quality that makes clothing look expensive — these Seedance 2.0 fashion prompts speak the visual grammar of high-end fashion video.
Fashion video is one discipline where the subject and the camera have to agree on their relationship before a single frame is generated. The model that carries a long tracking shot down a catwalk is different from the model that holds a static editorial frame while fabric settles — and the prompt has to specify which, or Seedance defaults to a generic walking clip that communicates nothing about the clothing, the brand, or the mood. The fashion prompts here are built around three core disciplines of fashion cinematography: fabric behavior, light quality, and camera-subject distance. Fabric behavior is the detail that separates fashion AI video from fashion AI video that works: specifying that a silk dress "catches the air as the model turns, the hem lagging half a beat behind, the light moving through the chiffon panels" gives Seedance a textile physics instruction rather than a color description. Light quality determines whether a garment reads as editorial or commercial: soft-box studio fill with a slight top rake is the lookbook lighting grammar; golden-hour rim light that blows out the background belongs to the campaign format; hard side-light casting a dramatic shadow across structured tailoring is the editorial-magazine aesthetic. Camera-subject distance encodes the fashion register: a wide tracking shot that shows the full body in motion is runway; a medium close-up from waist to collarbone as the model adjusts the collar is a detail shot that works for editorial; an extreme close-up on the hand resting against the fabric is the luxury brand format that says premium without saying anything. The prompts here cover all three disciplines — full runway sequences, editorial stills that move, detail fragment shots, campaign golden-hour portraits, and the backstage documentary register that has become a genre of its own. Each specifies what the fabric does, what the light does, and how far away the camera is when both happen. Copy a prompt, swap in the garment and colorway, and keep the fabric motion and light source intact.
More use cases
Frequently asked questions
What are the best AI video prompts for fashion video?
The best fashion prompts specify three things: what the fabric does (a silk hem lags behind a turn, a structured jacket stays rigid as the arms drop, chiffon panels catch the air), what the light does (soft editorial box fill, golden-hour rim, hard directional side-light for structured tailoring), and how far away the camera is (wide tracking for runway, medium close-up for detail, extreme close-up for luxury fragments). Without all three, Seedance defaults to a generic person wearing clothing rather than a fashion film moment. Every prompt in this gallery uses that three-layer structure.
How do I make runway AI video with Seedance 2.0?
Structure the runway prompt as a continuous tracking shot: camera follows at waist height from front-left, model walks toward camera at a deliberate pace, the full-length garment visible with the hem catching each step, the walk ending on a three-quarter turn and hold. Specify the lighting rig (soft overhead fill with a slight front rake, no harsh shadows on the garment surface) and the sound register (ambient catwalk sound with distant crowd, or silence for editorial). Naming the walk pace — deliberate, gliding, power stride — tells Seedance the energy level without requiring you to specify each step. Browse the fashion prompts here for full examples with preview videos.
Can Seedance 2.0 generate editorial fashion video?
Yes. Seedance 2.0 handles editorial fashion well when the prompt names the editorial register explicitly — 'editorial magazine style, static camera, model holds the opening stance then releases a slow breath and the jacket settles' — rather than asking for a fashion video generically. Editorial fashion video is defined by: a static or very slow push-in camera (the model moves, the camera barely does), a specific held pose at the start or end of the clip, and a defined light source that is directional enough to create shadow structure on the clothing. Add a location cue (brutalist concrete interior, minimal white studio, overgrown greenhouse) to complete the editorial frame.